From a2746b3f91740b6c98acbe960db9bb5895119a26 Mon Sep 17 00:00:00 2001
From: Christoph Urlacher <christoph.urlacher@protonmail.com>
Date: Tue, 1 Sep 2026 01:39:51 +0200
Subject: [PATCH] Add merged fault_probability chart

---
 .../combined_fault_probability_merged.r       | 117 ++++++++++++++++++
 1 file changed, 117 insertions(+)
 create mode 100644 scripts/charts/combined_fault_probability_merged.r

diff --git a/scripts/charts/combined_fault_probability_merged.r b/scripts/charts/combined_fault_probability_merged.r
new file mode 100644
index 0000000..cf19fe3
--- /dev/null
+++ b/scripts/charts/combined_fault_probability_merged.r
@@ -0,0 +1,117 @@
+library(ggplot2)
+library(dplyr)
+library(readr)
+library(stringr)
+library(tidyr)
+
+# Usage: Rscript combined_fault_probability_merged.r exp_abspath1 ... resultsdata_file
+
+args <- commandArgs(trailingOnly = TRUE)
+if (length(args) < 2) {
+  stop("Need at least 2 experiments")
+}
+
+csv_suffix <- if (grepl("\\.csv$", args[length(args)])) {
+  args[length(args)]
+} else {
+  "resultsdata.csv"
+}
+exp_args <- if (grepl("\\.csv$", args[length(args)])) {
+  args[-length(args)]
+} else {
+  args
+}
+
+extract_info <- function(path) {
+  dir_name <- basename(path)
+  match <- str_match(
+    dir_name,
+    "^\\d{2}-\\d{2}_\\d{2}-\\d{2}-\\d{2}_(.+?)_(c|aot|interp)_"
+  )
+  if (is.na(match[1, 1])) {
+    warning(paste("Could not parse:", dir_name))
+    return(NULL)
+  }
+
+  # "tacle-kernel-bsort" -> "bsort", otherwise it doesn't fit
+  base_name <- sub("^tacle-[^-]+-", "", match[1, 2])
+
+  list(base_name = base_name, variant = match[1, 3], path = path)
+}
+
+# Load data
+all_data <- data.frame()
+
+for (arg in exp_args) {
+  info <- extract_info(arg)
+  if (is.null(info)) {
+    next
+  }
+
+  csv_file <- file.path(info$path, csv_suffix)
+  if (!file.exists(csv_file)) {
+    warning(paste("Missing:", csv_file))
+    next
+  }
+  df <- read_csv(csv_file, col_types = cols())
+  df$base_name <- info$base_name
+  df$variant <- info$variant
+  all_data <- bind_rows(all_data, df)
+}
+
+# TODO: Finally put all the bullshit before this in some shared space
+
+if (nrow(all_data) == 0) {
+  stop("No data loaded")
+}
+
+marker_order <- c(
+  "OK_MARKER",
+  "DETECTED_MARKER",
+  "GROUP1_MARKER",
+  "TRAP",
+  "TIMEOUT",
+  "WRITE_TEXTSEGMENT",
+  "ACCESS_OUTERSPACE",
+  "FAIL_MARKER"
+)
+
+# Merge ip/mem/regs, then divide by the merged faultspace area
+# Don't merge GROUP1_MARKER into TRAP for this chart
+# Also keep the OK_MARKERs, so the "sum to 100%" is accurate
+probability <- all_data |>
+  group_by(base_name, variant, resulttype) |>
+  summarise(faults = sum(faults, na.rm = TRUE), .groups = "drop") |>
+  group_by(base_name, variant) |>
+  mutate(frac = faults / sum(faults, na.rm = TRUE)) |>
+  ungroup()
+
+# Don't print alphabetically
+probability$resulttype <- factor(probability$resulttype, levels = marker_order)
+probability$variant <- factor(
+  probability$variant,
+  levels = c("c", "aot", "interp")
+)
+
+plot <- ggplot(
+  probability,
+  aes(x = variant, y = frac, fill = resulttype)
+) +
+  geom_col() +
+  facet_wrap(~base_name) +
+  scale_y_continuous(labels = scales::percent) +
+  labs(
+    title = "Fault Probability (merged ip + mem + regs)",
+    x = NULL,
+    y = "Probability of Failure",
+    fill = "Fault Type"
+  ) +
+  theme_minimal() +
+  theme(
+    plot.title = element_text(size = 13, face = "bold"),
+    axis.text.x = element_text(angle = 45, hjust = 1)
+  )
+
+out_suffix <- gsub("^resultsdata|\\.csv$", "", csv_suffix)
+filename <- paste0("injections/fault_probability_merged", out_suffix, ".svg")
+ggsave(filename, plot = plot, width = 13, height = 8)